A first experiment using AI to write educational content

In this short article, I want to share with you how I've been mixing content generated from an AI with my own writing to create educational content about service design.

Generative AI is getting pretty good

As many lately, I've been pretty curious about what generative AI can do. My tool of choice Notion just launched an alpha version of its AI tools, I've been playing around with AI copywriting tools like Copy.ai, or Rytr and ChatGPT has proven to be a fun AI tool to ask questions to.

But you might ask, what is generative AI? Let me share with you what ChatGPT has to answer about this:


Generative AI is a type of artificial intelligence that is capable of generating new content based on data inputs that it has been trained on. This content can be in the form of text, images, audio, or other types of media.

There are several benefits to using generative AI:

  1. Increased efficiency: Generative AI can help to automate tasks that would otherwise be time-consuming or tedious for humans to complete.
  2. Improved accuracy: Generative AI can help to reduce errors and improve the accuracy of certain tasks by removing the possibility of human error.
  3. Enhanced creativity: Generative AI can help to generate novel and creative ideas that may not have been considered by humans.
  4. Increased scalability: Generative AI can help to scale certain tasks by allowing them to be completed faster and more efficiently.

However, there are also some limitations to using generative AI:

  1. Dependence on data: Generative AI relies on large amounts of data to learn from and generate new content. If the data is biased or incomplete, it can lead to biased or flawed outputs from the AI.
  2. Difficulty in explaining results: Generative AI can be difficult to understand and interpret, making it challenging to explain the results or decisions made by the AI to others.
  3. Limited to specific tasks: Generative AI is generally limited to specific tasks or areas of expertise and may not be able to perform tasks outside of its training data.
  4. Ethical concerns: There are also ethical concerns surrounding the use of generative AI, including the potential for it to be used for nefarious purposes or to replicate or create harmful content.


What it's good at

In my experiments, I've found that these AI tools are pretty good at summarising basic knowledge that is already well-documented online. 

AI to write educational content about service design

So now, when I'm trying to answer a service design question that I know already has pretty good answers all over the internet, I do this:
  1. I write my personal opinion: and focus only on what's different in my opinion
  2. I ask an AI like ChatGPT the question: and this helps me get a summary of the general ideas

Asking ChatGPT "What are good reasons to build a service blueprint?"

This allows me to have both the general answer that would be easy to google but formatted in a good summary and an opinionated and human answer that feels like the extra sauce that only human experience can provide.

So far, I've played with this idea on the two following service design questions and answers:
  1. What's the definition of a Service Blueprint?
  2. Why should I create a service blueprint?

This blog post is also a pretty fun experiment that mixes my own writing that shares my personal experience and the generative writing of an AI that explains common knowledge.

I'll continue to play with these tools in the future by sharing clearly what is human-based and what not, as I feel this offers a lovely look into the backstage for the curious mind.

Written on December 17th 2022 by Daniele Catalanotto